Industrial closed culture of Marine microalgae requires higher environmental parameters in the whole process.Synechococcus PCC7002 was selected as the culture object in the pilot stage of the culture, and a variety of parameters in the culture process were collected and a data set was established. A growth prediction model of synechococcus PCC7002 based on LSTM neural network was constructed. Mul_x005ftiple environmental parameters, such as temperature, pH, light intensity, air input and dissolved oxygen value, were used as input parameters of the model. After repeated training and learning, turbidity value (i.e., cell concentration) of the growth state of microalgae was obtained. The turbidity value of quantifiable microalgae growth conditions and the growth trend of PCC7002 were obtained after the coupled training of the main environmental factors in the microalgae culture process. Then, the MTS-Mixers algorithm was integrated to predict the external environment prediction curve required for the growth of microalgae in the stable and exponential stages in the culture process.